Hi I have two parquet files they are made by different way,
1st Use pig load protobuf file then store by parquet pig storer(parquet-pig-bundle 1.9.0) 2nd Use protobuf to parquet and add this patch(https://github.com/apache/parquet-mr/pull/411) or store by spark But when I use parquet pig loader to read 2nd example and describe schema, it show different schema with I use same loader to read 1st case(read example and describe). And when read 2nd example then access tuple I can't use "." to point out tuple content like normal pig script, error message is "Cannot find field name in element:tuple(name:chararray,name2:chararray)”. *read example and describe: 1st case: value:(guid: chararray, blob: (info: {info_tuple: (name: chararray,name2: chararray)}) 2nd case: value:(guid: chararray, blob: (info: {list: (element: (name: chararray,name2: chararray))})) *access tuple: 1st case: B = foreach A generate value.blob.info.name; 2nd case: B = foreach A generate value.blob.info.name; //error message: Cannot find field name in element:tuple(name:chararray,name2:chararray) And I ask Julien, when use 2nd case I can add “.element.” like “ B = foreach A generate value.blob.info.element.name;” to get tuple value, but element is a virtual level, like “ info_tuple”, maybe need add some solution to avoid virtual level Thank you and regards, Abel ke <table class="TM_EMAIL_NOTICE"><tr><td><pre> TREND MICRO EMAIL NOTICE The information contained in this email and any attachments is confidential and may be subject to copyright or other intellectual property protection. If you are not the intended recipient, you are not authorized to use or disclose this information, and we request that you notify us by reply mail or telephone and delete the original message from your mail system. </pre></td></tr></table>
